class: center, middle, inverse, title-slide .title[ # Survey Data Analysis with Kobocruncher ] .subtitle[ ## Session 2 - Relabeling ] .author[ ### <a href="https://edouard-legoupil.github.io/kobocruncher/">Link to Documentation</a> – <a href="01-First_Report.html">Link to Previous Session</a> – <a href="03-Grouping_Questions.html">Link to Next Session</a> ] .date[ ### Training Content as of 29 November 2022 ] --- ## Why do you need to relabel your dataset When you obtain the raw data from the kobo server, the questions labels are the one used for data collection The way questions are phrased during the interview might not be optimal for the data analysis stage --- ## What is a good question and response label for data analysis? Label should be short and concise in order to ease quick reading (capture the content of the final chart in a few seconds) Ideally: * Less than 80 characters for variable labels * Less than 40 characters for responses Ideally try to summarize the initial wording --- ## Other issues to fix when relabeling In case of skip logic questions, --- ## --- class: inverse, center, middle # TIME TO PRACTISE ON YOUR OWN 5 minutes! - Expand the xlsfom - Download locally and adjust the wording of the questions - upload and knit again your report --- class: inverse, center, middle # Thank you __Next session__: [03-Grouping_Questions](03-Grouping_Questions.html) Within a questionnaire, questions follows a certain sequence in order to facilitate the interview. When exploring the results, it is often required to regroup the questions in a different way and inline with key research questions